
% Table created by stargazer v.5.2.3 by Marek Hlavac, Social Policy Institute. E-mail: marek.hlavac at gmail.com
% Date and time: Wed, Jul 19, 2023 - 11:37:05
% Requires LaTeX packages: dcolumn 
\begin{table}[!htbp] \centering 
  \caption{Spatial Durbin Error Models on support for VOX} 
  \label{tab:sdem} 
\small 
\begin{tabular}{@{\extracolsep{-20pt}}lD{.}{.}{-3} D{.}{.}{-3} D{.}{.}{-3} } 
\\[-1.8ex]\hline 
\hline \\[-1.8ex] 
\\[-1.8ex] & \multicolumn{1}{c}{(1)} & \multicolumn{1}{c}{(2)} & \multicolumn{1}{c}{(3)}\\ 
\hline \\[-1.8ex] 
 (Intercept) & -0.260^{***} & -0.332^{***} & 11.088^{***} \\ 
  & (0.012) & (0.010) & (1.860) \\ 
  Military facility & 0.022^{***} & 0.027^{***} & 0.038^{***} \\ 
  & (0.002) & (0.003) & (0.003) \\ 
  Military (spatial lag) & 0.017^{**} & 0.019 & 2.441^{***} \\ 
  & (0.005) & (0.013) & (0.121) \\ 
 \hline \\[-1.8ex] 
Controls & \multicolumn{1}{c}{Yes} & \multicolumn{1}{c}{Yes} & \multicolumn{1}{c}{Yes} \\ 
Lambda & \multicolumn{1}{c}{0.89$^{***}$} & \multicolumn{1}{c}{0.90$^{***}$} & \multicolumn{1}{c}{1.00$^{***}$} \\ 
Observations & \multicolumn{1}{c}{20,181} & \multicolumn{1}{c}{20,181} & \multicolumn{1}{c}{20,181} \\ 
Akaike Inf. Crit. & \multicolumn{1}{c}{-89,573.300} & \multicolumn{1}{c}{-81,628.430} & \multicolumn{1}{c}{-79,812.380} \\ 
\hline 
\hline \\[-1.8ex] 
\multicolumn{4}{c}{\parbox[t]{0.6\textwidth}{\textit{Note:} $+ p<0.1; * p<0.05; ** p<0.01; *** p<0.001$. Models 1 includes spatial weights based on queen-type contiguity. Model 2 does so identifying neighbors as sections within 2km. Model 3 uses spatial weights based on the inverse logged distance (m). Controls: turnout, logged section population, logged household income, logged municipality population, dummy for military region capital. Full table in Appendix.}} \\ 
\end{tabular} 
\end{table} 
